fitness-report

craigm26/fitness-report

★ 0 stars TypeScript AI/LLM Updated today

Find out whether agents can actually drive your MCP server before your customers do. Scored runs with validity gates that refuse to score an invalid eval.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "fitness-report": {
      "command": "npx",
      "args": [
        "-y",
        "craigm26/fitness-report"
      ]
    }
  }
}

README Excerpt

Find out whether agents can actually drive your MCP server before your customers do. One scored run against one server version. You get first-try task success with a confidence interval, calls and tokens per completed task, per-tool failure attribution, suggested tool-description rewrites grounded in recorded failures,

fitness-report

Quick Install

README Excerpt

Related AI/LLM Servers